Bellis, C., Nobbs, A., O'Sullivan, D., Holder, J., & Barbour, M. (2016). Glassionomer cements functionalised with a concentrated paste of chlorhexidinehexametaphosphate provides dose-dependent chlorhexidine release over atleast 14 months. Journal of Dentistry, 45, 53-58. DOI:10.1016/j.jdent.2015.12.009

Peer reviewed version

Link to published version (if available):10.1016/j.jdent.2015.12.009

Link to publication record in Explore Bristol ResearchPDF-document

University of Bristol - Explore Bristol ResearchGeneral rights

This document is made available in accordance with publisher policies. Please cite only the publishedversion using the reference above. Full terms of use are available:http://www.bristol.ac.uk/pure/about/ebr-terms.html

http://dx.doi.org/10.1016/j.jdent.2015.12.009http://research-information.bristol.ac.uk/en/publications/glass-ionomer-cements-functionalised-with-a-concentrated-paste-of-chlorhexidine-hexametaphosphate-provides-dosedependent-chlorhexidine-release-over-at-least-14-months(9f27f6de-29af-4c00-946e-4482a53b0932).htmlhttp://research-information.bristol.ac.uk/en/publications/glass-ionomer-cements-functionalised-with-a-concentrated-paste-of-chlorhexidine-hexametaphosphate-provides-dosedependent-chlorhexidine-release-over-at-least-14-months(9f27f6de-29af-4c00-946e-4482a53b0932).html

1

Glass ionomer cements functionalised with a concentrated paste of chlorhexidine

hexametaphosphate provides dose-dependent chlorhexidine release over at least 14 months

Abstract

Objectives

The aim of this study was to create prototype glass ionomer cements (GICs) incorporating a

concentrated paste of chlorhexidine-hexametaphosphate (CHX-HMP), and to investigate the long-

term release of soluble chlorhexidine and the mechanical properties of the cements. The purpose is

the design of a glass ionomer with sustained anticaries efficacy.

Methods

CHX-HMP paste was prepared by mixing equimolar solutions of chlorhexidine digluconate and sodium

hexametaphosphate, adjusting ionic strength, decanting and centrifuging. CHX-HMP paste was

incorporated into a commercial GIC in substitution for glass powder at 0.00, 0.17, 0.34, 0.85 and 1.70%

by mass CHX-HMP. Soluble chlorhexidine release into artificial saliva was observed over 436 days using

absorbance at 255 nm. Diametral tensile and compressive strength were measured after 7 days’

setting (37°C, 100% humidity) and tensile strength after 436 days’ aging in artificial saliva. 0.34% CHX-

HMP GICs were tested for their ability to inhibit growth of Streptococcus mutans in vitro.

Results

GICs supplemented with CHX-HMP exhibited a sustained dose-dependent release of soluble

chlorhexidine. Diametral tensile strength of new specimens was unaffected up to and including 0.85%

CHX-HMP, and individual values of tensile strength were unaffected by aging, but the proportion of

CHX-HMP required to adversely affect tensile strength was lower after aging, at 0.34%. Compressive

strength was adversely affected by CHX-HMP at substitutions of 0.85% CHX-HMP and above.

Conclusions

2

Supplementing a GIC with CHX-HMP paste resulted in a cement which released soluble chlorhexidine

for over 14 months in a dose dependent manner. 0.17% and 0.34% CHX-HMP did not adversely affect

strength at baseline, and 0.17% CHX-HMP did not affect strength after aging. 0.34% CHX-HMP GICs

inhibited growth of S. mutans at a mean distance of 2.34 mm from the specimen, whereas control

(0%) GICs did not inhibit bacterial growth.

Clinical Significance

Although GICs release fluoride in vivo, there is inconclusive evidence regarding any clinical anticaries

effect. In this study, GICs supplemented with a paste of chlorhexidine-hexametaphosphate (CHX-HMP)

exhibited a sustained release of chlorhexidine over at least 14 months, and small additions of CHX-

HMP did not adversely affect strength.

Keywords: glass ionomer; restorative; antimicrobial; chlorhexidine; biomaterials

3

Introduction

Glass ionomer cements (GICs) are used for a number of purposes, including as direct restorative

materials, lining and luting materials, adhesives, and in atraumatic restorative therapy. GIC

restorations typically have shorter lifetimes than composites or amalgams (1-3), although the reasons

for this are complex and, of course, the materials are not selected at random but are chosen by the

clinician according to clinical need. When a GIC does fail, there are a number of potential reasons, one

of which is secondary caries; this is responsible for 25% of failures of GIC-lined restorations after 18

years of clinical service (4) and around 18% of GIC failures over a range of 0.1-23 years (5).

GICs leach fluoride into the oral environment. This results in elevated fluoride concentrations close to

the restoration, and thus there is an hypothesis that this may reduce dental caries in the local area

owing to the interaction of the fluoride ion with the hydroxyapatite in the enamel and dentine. This

hypothesis is broadly supported by in vitro data, but in situ and clinical studies of caries incidence in

the vicinity of fluoride-releasing restoratives do not show consistent results (6). At the current time it

is not possible to conclude whether fluoride release from GICs provides lasting protection against

dental caries (6, 7).

Chlorhexidine (CHX) is a biocide with broad spectrum efficacy against a wide range of microbes. Its

main application in dentistry is as a topical agent, usually in oral care products, products for treatment

of periodontal disease, and varnishes. CHX is efficacious against the microbes implicated in dental

caries, and is used in a number of products designed to protect the dentition against decay. However,

the CHX salts in currently available commercial form have poor retention in situ, providing typically a

few hours of antimicrobial function. One commercial material used for treatment of periodontal

disease provides some sustained CHX release, but this is a short-term effect with 80% of the CHX

released within the first 2 days, and a very slow release over the following 3-4 weeks (8).

4

There have been a number of attempts to incorporate CHX into GICs, with the aim of creating a

restorative material that offers lasting protection against caries. GICs doped with CHX diacetate and

CHX digluconate have been reported, and these inhibited growth of Streptococcus mutans and

Lactobacillus acidophilus, but there was also some deterioration of mechanical properties and the

antimicrobial effects were limited to the first 40-90 days of the study, with no bactericidal effect

observed after this time (9). There are also reports of an increase in porosity and setting time and a

reduction in hardness and tensile bond strength when GICs are doped with CHX digluconate (10, 11).

CHX diacetate and CHX hydrochloride have also been incorporated into GICs, and these too inhibited

growth of caries-causing organisms, but CHX release was only observed for 24 h so it is not clear how

long this effect would be sustained (12). CHX diacetate doped resin modified GICs exhibited some

sustained release of CHX, although this reached completion in 14-21 days (13). The release profiles of

soluble CHX from GICs doped with these conventional salts of CHX exhibit a high initial release

followed by little or no sustained release, and this perhaps explains the cytotoxic effects observed

against fibroblasts when GICs were doped with 1 % CHX diacetate (14). However, supplementation of

a resin-modified GIC with CHX digluconate at modest concentrations (1.25%) had no adverse effects

on osteoblasts in vitro and resulted in an elimination of Streptococcus mutans populations following

indirect pulp treatment in vivo (15), suggesting a potential clinical benefit of a CHX-functionalised

restorative material.

We have previously reported the use of a novel salt of CHX: CHX-hexametaphosphate (CHX-HMP) (16).

CHX-HMP has a lower solubility than CHX digluconate or CHX diacetate and, when used as a coating

or dopant, can confer a sustained release of CHX that persists for at least three months (17). We have

described the use of CHX-HMP as a filler for GICs (18). In that study, large clusters of CHX-HMP particles

were used, and the size of these large particles, which were formed due to the production process, is

likely to account for the adverse effects on the mechanical properties observed. The aim of the study

described here was to investigate the use of CHX-HMP particles as GIC fillers but using a new

preparation method which omits the drying process which creates the large aggregates and instead

5

uses a process of ionic strength adjustment and centrifugation to sequester the particles. CHX release

was probed over a clinically relevant timescale of over one year, and both compressive and tensile

strength were investigated; the latter was measured also after 14 months’ aging to determine if the

modification of the cement adversely affects long-term mechanical properties.

The hypothesis was that the prototype cements incorporating a concentrated paste of CHX-HMP

particles would confer a more sustained CHX release, sufficient to inhibit growth of cariogenic

microbes, coupled with less adverse effects on mechanical properties in comparison to large, dry

aggregates of CHX-HMP or conventional salts of CHX such as digluconate or diacetate.

Methods

Preparation of CHX-HMP paste

Aqueous 10 mM solutions of CHX digluconate and sodium HMP were prepared. 100 mL of each

solution were combined in a glass beaker under ambient laboratory conditions. The suspension

created was stirred vigorously for approximately 1 min, then 30 mL 1 M potassium chloride was added.

Stirring continued for a further 1 min before the preparation was allowed to settle for 24 h. The

precipitate settled at the bottom of the flask and the supernatant was gently discarded leaving a

concentrated suspension of the precipitate. This suspension was then centrifuged at 4760 g for 30

min. The supernatant was again discarded and the pellet of paste was removed from the centrifuge

tubes using a spatula and used immediately.

Preparation of specimens

A commercially available GIC, Diamond Carve™ (Kemdent Ltd, Purton, UK), was used as the base

material to create the experimental cements. This GIC comprises a powder, consisting of alumina-

silica based glass filler particles which contain calcium fluoride and other minor salt components and

freeze dried poly(vinyl)phosphonic acid, and a liquid, which contains polyacrylic and tartaric acids. The

6

manufacturers’ instructions indicate that the powder and liquid should be mixed in a 4:1 ratio by mass

to create the finished cement.

The water content of the paste was established to allow the concentration of the liquid component of

the GIC to be adjusted to account for the additional water in the CHX-HMP paste. CHX-HMP paste was

weighed as freshly prepared, then stored at 37oC and weighed periodically until the mass of the

powder was constant, indicating that the available water had evaporated (24 h). This revealed a

composition of 83% water and 17% CHX-HMP particles. The GIC liquid component was thus prepared

at a concentration that resulted in the standard final concentration when diluted by the paste. The

paste was substituted for the overall mass at 0, 0.17, 0.34, 0.85 and 1.70 % by mass of CHX-HMP (0,

1, 2, 5 and 10% by mass of the paste). The paste was mixed into the liquid first and then the powder

was added to the paste-liquid combination. Mixing of the specimens was completed in 40-50 s and

packing into the moulds took a further 10 s, such that all manipulation of the cement was completed

within 1 min.

GICs were packed, using a stainless steel spatula, into stainless steel moulds with dimensions of 6 mm

height and 4 mm diameter (for compressive strength determination) or 4 mm height and 6 mm

diameter (for measurement of diametral tensile strength and elution of CHX). The moulds were lined

with a thin layer of petroleum jelly to aid removal of the set cement. Immediately after packing, the

moulds were placed between two sheets of acetate and a 2 kg weight placed on top of the specimens

on a flat surface in order to ensure even distribution of the cement. After 5 minutes the specimens

were sanded using a P120 grit sanding disc (Hermes, Hamburg, Germany) to remove excess material

and were then placed into small, sealed plastic vessels containing wet tissue paper packed into the lid

to achieve 100% humidity without direct contact with water. Specimens were stored at 37oC for 7 days

prior to any further testing to ensure the setting process had reached completion.

7

Compressive strength (CS) testing

CS was measured by applying a compressive force to the flat surface of the cylindrical specimens using

a universal testing machine (Zwick/Roell Universal Testing Machine, Zwick/Roell, Leominster,

Herefordshire, UK) and recording the load at fracture. Specimens were examined after fracture for

evidence of flaws on the internal or external surfaces and data from flawed specimens were rejected.

Load at fracture LF was used with diameter D to calculate CS according to the relationship CS = 4L/πD2.

The load was used in conjunction with the average diameter of the specimen. Specimen dimensions

were measured three times using a digital micrometer. N = 24 specimens per group were used. Data

were analysed using a one-way ANOVA followed by a Tukey Honestly Significant Difference post-hoc

test.

Diametral tensile strength (DTS) testing

DTS was measured by applying a compressive force to the curved sides of the cylindrical specimen

using a universal testing machine and recording the load at fracture. Specimens were examined after

fracture for evidence of flaws on the internal or external surfaces and data from specimens found to

be flawed were rejected. The load at fracture LF was used in conjunction with the average diameter D

and height h of the specimens to calculate DTS according to the relationship DTS = 2L/πDh. Specimen

dimensions were measured three times using a digital micrometer.

N = 24 specimens per group were used for the “new” specimens; these were prepared, allowed to set

for 7 days then tested immediately. Specimens were also tested after 436 days’ elution, to establish

the diametral tensile strength of “aged” specimens. N = 15 specimens were used for this test. Data

were analysed using one-way ANOVA followed by Tukey Honestly Significant Difference post-hoc

tests.

8

Characterisation of CHX release

GIC specimens were weighed using a precision balance then placed in individual cuvettes (Z637157-

100EA, Sigma-Aldrich, Gillingham, UK) transparent to ultraviolet (UV) light. 1.5 mL artificial saliva was

added to each cuvette. The artificial saliva was composed of 0.9 mM CaCl2, 0.2 mM MgCl2, 4.0 mM

KH2PO4, 30.0 mM KCl, 20.0 mM HEPES buffer, titrated to pH of 6.8. The cuvettes were sealed with

tightly-fitting lids (SEMA2533, VWR, Lutterworth, UK) and then placed onto an orbital shaker (SSM1,

Stuart, Staffordshire, UK) at 150 rpm and readings taken initially once a day, and less frequently as

CHX release decelerated. Artificial saliva was refreshed at two-week intervals to avoid any decrease in

CHX release that could be attributed to saturation of the artificial saliva with respect to CHX salts.

Adsorption of light at wavelength 255 nm was measured at regular intervals using a

spectrophotometer (Hitachi U-1900, Hitachi, Japan) and calibration standards of 5 – 50 µM CHX used

as references to establish CHX release from the GICs into the artificial saliva (19). This was converted

to μmoles CHX released per unit surface area for each specimen and normalised by subtracting the

mean reading for the 0% substitution, correcting for other eluents of the GIC such as the polyacrylic

acid (18). N = 15 specimens per group were used.

Microbiological testing

Streptococcus mutans GS-5 was cultivated anaerobically at 37 oC on BHYN agar (per litre: 37 g Brain

Heart Infusion, 5 g yeast extract, 5 g Neopeptone, 15 g agar). Suspension cultures were grown in BHY

medium (per litre: 37 g Brain Heart Infusion, 5 g yeast extract) in sealed bottles and incubated

stationary at 37 oC for 16 h. Bacterial cells were washed once with phosphate-buffered saline (PBS) by

alternate centrifugation (5000 g, 7 min) and suspension, and suspended in PBS at OD600 1.0

(approximately 2 x 108 cells/ml). A lawn of bacterial growth was generated by spreading 100 l of

adjusted S. mutans suspension onto a BHYN agar plate, which was then incubated anaerobically at 37

oC for 16 h.

9

The GIC specimens were made using the method described above in a 0% (control) and a 0.34%

substitution. These specimens were prepared using a sterile spatula and moulds in a biological safety

cabinet (ESCO airstream, ESCO micro pte ltd, Singapore). The specimens were left to mature in a moist

environment for 7 days at 37 oC, then placed onto the bacterial lawn plates (with minimal force to

ensure no movement once in the incubator). Lawn plates were incubated for 16 h at 37 oC under

anaerobic conditions and the zones of inhibition (clearance) measured by diameter.

Results

Compressive strength

Rejection of specimens with internal voids, imperfections or non-linear force-distance curves resulted

in final n per specimen group of 16-21 (mean n = 19). CS data are shown in Figure 1. There was no

statistically significant difference between CS of control, 0.17 or 0.34% CHX-HMP specimens. 0.85 CHX-

HMP had significantly lower CS than control, 0.17 or 0.34% specimens, and 1.70% CHX-HMP had

statistically significantly lower CS than all other groups.

Diametral tensile strength

Rejection of specimens with internal voids, imperfections or non-linear force-distance curves resulted

in final n per specimen group of 18-24 (mean n = 21). DTS data are shown in Figure 2. The only group

that was statistically significantly different from the control was the 1.70 % CHX-HMP, which had a

reduced DTS (p

10

values, but these were not statistically significant. For the aged specimens, 0% and 0.17% formed a

homogeneous group, 0.34% was significantly lower than 0 and 0.17%, and 0.85% and 1.70% were

significantly lower still.

Characterisation of CHX release

Elution of soluble CHX from CHX-HMP doped GICs normalised to control specimens are shown in

Figure 4 and Figure 5. Figure 4 shows mean cumulative CHX elution for specimens with 1.70% CHX-

HMP at all time points measured. The purpose of displaying the data in this way is to illustrate the

smooth curve, indicating that the practice of refreshing the artificial saliva elution medium at 14 day

intervals was sufficient to prevent saturation with respect to CHX salts from inhibiting the release of

further CHX. There is a slight suggestion that there is a step in CHX concentration between day 28 and

day 29 (ie before and after a saliva change) but no other such step changes or accelerations of CHX

elution are observed.

In Figure 5, data from day 14x is shown illustrating CHX release as a function of time and dose of CHX-

HMP. A dose response is observed, with greater CHX-HMP substitution in the GIC resulting in a greater,

and faster, release of CHX-HMP throughout the experimental period. All specimens groups were still

releasing CHX at the conclusion of the experiment, with the greatest release observed for the higher

CHX-HMP substitutions.

The mean cumulative CHX release at each 14x day time was compared to the lowest substitution, to

establish the relationship between CHX-HMP dose and CHX release. There was a non-linear

relationship between CHX-HMP dose and CHX release; while the ratio of CHX-HMP concentration for

0.17, 0.34, 0.85 and 1.70 % CHX-HMP was 1:2:5:10, the ratio of CHX release was 1:4.6:12.3:30.6.

11

Microbiological testing

0.34% CHX-HMP GICs exhibited zones of inhibition on the lawn of Streptococcus mutans. The mean

zone of inhibition was 2.34 mm (standard deviation 1.19 mm) from the periphery of the specimen.

The control GICs displayed no zones of inhibition.

Discussion

CHX-HMP particles in a concentrated aqueous paste were prepared and incorporated into a

commercial GIC, using a bespoke polyacid formulation to correct for the water fraction of the CHX-

HMP paste. The experimental GICs exhibited a sustained release of aqueous CHX into artificial saliva.

This CHX release continued for the duration of the experiment (436 days), and although all decelerated

over the course of the experiment, even the lowest substitution (0.17%) had a gradient >0 at the

conclusion of the experiment. For the higher substitutions there was still a steady and substantial

release of CHX when the experiment concluded.

The CHX release observed from GICs supplemented with CHX-HMP was sustained for much longer

than that achieved using CHX digluconate or CHX diacetate as the dopant (9). This is likely to be owing

to the low solubility of CHX-HMP; the CHX salt is added as a solid filler which is incorporated into the

GIC structure, only releasing its soluble CHX payload as some of the particles gradually dissociate. This

is in contrast to the approach of supplementing a GIC with CHX digluconate or diacetate; in this case

the CHX is added either already in solution (digluconate) or as a solid, but highly soluble, filler. It is not

surprising that these approaches lead to a rapid release of CHX in contrast to CHX-HMP.

CHX release was dose-dependent, with a non-linear relationship; the greater the CHX-HMP in the GIC,

the more the CHX release, but CHX release increased faster than CHX-HMP dose. This may reflect a

less stable structure for the higher substitutions; with the higher dopings of CHX-HMP, the particles

disrupt the setting process of the GIC, meaning that the GIC is more porous and the CHX-HMP particles

embedded deep within the cement structure are “accessible” to the artificial saliva and contribute to

12

the CHX release. In contrast, those with lower dopings of CHX-HMP have an effective setting process

and the particles that contribute to the CHX release are only those close to the GIC’s external surface.

This is supported by the mechanical property data (discussed below) which indicates that the higher

dopings of CHX-HMP adversely affect the material’s strength.

The presence of soluble CHX in the elution medium was insufficient during the experiments to restrict

any further CHX release owing to the large solution:surface area ratio, the agitation of the reaction

vessels (cuvettes) and the regular changes of artificial saliva. It should be noted that the conditions

used in this study are not an accurate representation of the clinical scenario; much higher shear

conditions and larger volumes of saliva were used in this study than would be in contact with the GIC

at the margins of a restoration. As such it is likely that the rate of CHX release would be slower in vivo,

because local low shear forces and small fluid volumes would result in a local saturation with respect

to CHX. That is to say, it is likely that CHX concentrations at the restoration-tooth interface would be

higher than reported here, as there is little shear or diffusion to remove the CHX, and that these locally

high CHX concentrations would likely slow the subsequent release of CHX owing to saturation effects.

Thus the true duration of CHX release from these cements could only be determined from a study

with a more accurate depiction of the clinical conditions, but it is hypothesised to be longer than that

observed here.

DTS of the GICs was not statistically significantly affected by addition of CHX-HMP, compared to the

unmodified cement, up to and including 0.85% CHX-HMP (Figure 2). There was a numerical rise in DTS

up to 0.34% CHX-HMP and then a fall with the lowest value recorded for 1.70% CHX-HMP, but only

the 1.70% CHX-HMP group differed from the control to a statistically significant degree. DTS of

specimens aged for 436 days indicated no statistically significant changes compared to the baseline

values. This is consistent with previously published data, indicating that while flexural strength of GICs

typically decreases after laboratory aging, by as much as 55% (20), compressive and tensile strengths

tend to remain similar or increase (21), as was observed here. Although there were increases in DTS

13

for control and 0.17 % CHX-HMP and a decrease in DTS for 0.34% CHX-HMP, these were not

statistically significant (Figure 3). One outcome of this was that the substitution of CHX-HMP required

to affect a reduction in DTS was lower with aged specimens; 0.34% CHX-HMP specimens had a

statistically significantly lower DTS than control specimens after 436 days’ aging, as compared to

1.70% for new specimens. This development can be attributed to the increase in DTS for control

specimens, as well as the decrease in DTS for specimens with 0.34% CHX-HMP, associated with aging.

CS of GICs was affected by addition of CHX-HMP; cements with 0.84 and 1.70% CHX-HMP had

statistically significantly reduced CS compared with control, but those with 0.17 and 0.34% were

equivalent to the unmodified cement. CS is acknowledged as one of the more sensitive properties as

regards modifications to GIC formulations (22) and thus it is not surprising that this parameter was

affected by CHX-HMP addition more than DTS.

The lower strengths of the GICs with the higher substitutions of CHX-HMP can be explained by

observing that, in increasing the doping of CHX-HMP, the ratio of polyacid to glass is altered, with less

glass to account for the greater proportion of CHX-HMP. This will disrupt the GIC setting process as

fewer glass particles and thus fewer bi- and trivalent ions are available to cross-link the polyacid

matrix. Thus those specimens with more CHX-HMP are likely to have a less complete setting reaction,

and this is likely to explain the reduced strength for these specimens.

The 0.34% CHX-HMP GIC was tested for its ability to inhibit the growth of oral pathogen Streptococcus

mutans in vitro. This GIC formulation was selected on the basis that it was the highest substitution of

CHX-HMP that did not have an adverse effect on initial mechanical properties. It was observed that

these GICs inhibited growth of the microorganism to a mean distance of 2.34 mm from the periphery

of the specimen, whereas control GICs displayed no zone of inhibition. This indicates that, in vitro and

utilising a simple, single species model, the concentration of CHX released by the 0.34% CHX-HMP GIC

was sufficient to inhibit growth of this cariogenic organism.

14

The strength data, particularly the DTS of the aged specimens combined with the CS data, suggests

that if such a GIC were to find clinical application, the lower substitutions of CHX-HMP paste would be

more suitable, as these have no negative impact on mechanical properties. The 0.34% CHX-HMP GIC,

which was the highest concentration of CHX-HMP while not adversely affecting mechanical properties

at baseline, inhibited growth of Streptococcus mutans in an in vitro zone of inhibition model. Noting

the comments above regarding the experimental conditions, the next step should be to determine

whether this cement can inhibit the growth of caries-causing microorganisms under flow conditions

more representative of the clinical environment of a GIC.

Conclusions

CHX-HMP particles in a concentrated paste were incorporated into a commercial GIC, with adjustment

of the acid concentration in the commercial GIC to account for the water in the paste. Higher

substitutions of CHX-HMP paste (0.85 and 1.70% CHX-HMP by mass, or 5 and 10% paste by mass) had

a negative impact on compressive strength, but lower substitutions had no significant impact on

compressive strength. None of the prototype cements displayed a reduction in diametral tensile

strength at baseline; DTS after 436 days aging was not statistically significantly different when

comparing specimens with and without aging, although the substitutions of 0.34% and higher were

significantly weaker than the control cement after aging. All of the prototype cements exhibited a

sustained release of soluble CHX over the 436 day experimental period.

15

Figures

Figure 1. CS of GIC specimens as a function of CHX-HMP substitution. Error bars represent

standard deviations.

0

20

40

60

80

100

120

140

160

180

0.00 0.17 0.34 0.85 1.70

Co

mp

ress

ive

stre

ngt

h [

MP

a]

% substutition of CHX-HMP

16

Figure 2. DTS of GIC specimens as a function of CHX-HMP substitution. Error bars represent

standard deviations.

0

2

4

6

8

10

12

0.00 0.17 0.34 0.85 1.70

Dia

me

tral

te

nsi

le s

tre

ngt

h [

MP

a]

% substitution of CHX-HMP

17

Figure 3. Diametral tensile strength of newly prepared GIC specimens (“new”) and specimens

immersed in artificial saliva, refreshed fortnightly, for 436 days (”aged”). p values refer to

comparisons of new and aged specimens for a given % substitution of CHX-HMP; although

numerical differences were observed these were not statistically significant.

0

2

4

6

8

10

12

0 0.17 0.34 0.85 1.7

Dia

met

ral t

ensi

le s

tren

gth

[M

Pa]

% substitution of CHX-HMP

New

Aged

p=0.475p=0.201

p=0.208

p=1.000

p=0.997

18

Figure 4. Cumulative CHX release from GIC specimens containing 1.70% CHX-HMP in

substitution for the powder component. This figure shows data points from all sampling days,

indicating the smooth transition between the 14 day periods, and no local saturation existing

prior to artificial saliva change, with one possible exception where there is a step change from

day 28 (before saliva change) to day 29 (after saliva change) indicating the possibility of some

minor inhibition of CHX release at this single time point.

0

50

100

150

200

250

300

350

400

450

0 50 100 150 200 250 300 350 400 450

Cu

mu

lati

ve C

HX

rel

ease

[µ

mo

l.m-2

]

Time [days]

19

Figure 5. Cumulative CHX release from GIC specimens as a function of CHX-HMP substitution

for the powder component of the cement. Data is presented at 14 day intervals.

-50

0

50

100

150

200

250

300

350

400

0 50 100 150 200 250 300 350 400 450

Cu

mu

lati

ve C

HX

rel

ease

[µ

mo

l.m-2

]

Time [days]

0.17%

0.34%

0.85%

1.70%

20

Bibliography

1. Pinto Gdos S, Oliveira LJ, Romano AR, Schardosim LR, Bonow ML, Pacce M, Correa MB,

Demarco FF, Torriani DD. Longevity of posterior restorations in primary teeth: results from a

paediatric dental clinic. Journal of Dentistry 2014; 42: 1248-54.

2. Burke FJ, Lucarotti PS. How long do direct restorations placed within the general

dental services in England and Wales survive? British Dental Journal 2009; 206: E2; discussion

26-7.

3. Hurley E, Da Mata C, Stewart C, Kinirons M. A study of primary teeth restored by

intracoronal restorations in children participating in an undergraduate teaching programme

at Cork University Dental School and Hospital, Ireland. European Journal of Paediatric

Dentistry 2015; 16: 78-82.

4. van de Sande FH, Rodolpho PA, Basso GR, Patias R, da Rosa QF, Demarco FF, Opdam

NJ, Cenci MS. 18-year survival of posterior composite resin restorations with and without

glass ionomer cement as base. Dental Materials 2015; 31: 669-75.

5. Namgung C, Rho YJ, Jin BH, Lim BS, Cho BH. A retrospective clinical study of cervical

restorations: longevity and failure-prognostic variables. Operative Dentistry 2013; 38: 376-85.

6. Wiegand A, Buchalla W, Attin T. Review on fluoride-releasing restorative materials--

fluoride release and uptake characteristics, antibacterial activity and influence on caries

formation. Dental Materials 2007; 23: 343-62.

7. Yengopal V, Harneker SY, Patel N, Siegfried N. Dental fillings for the treatment of caries

in the primary dentition. Cochrane Database of Systematic Reviews 2009: CD004483.

21

8. Tabary N, Chai F, Blanchemain N, Neut C, Pauchet L, Bertini S, Delcourt-Debruyne E,

Hildeband HF, Martel B. A chlorhexidine-loaded biodegradable cellulosic device for

periodontal pockets treatment. Acta Biomateriala 2014; 10: 318-29.

9. Turkun LS, Turkun M, Ertugrul F, Ates M, Brugger S. Long-term antibacterial effects

and physical properties of a chlorhexidine-containing glass ionomer cement. Journal of

Esthetics and Restorative Dentistry 2008; 20: 29-44; discussion 45.

10. Marti LM, Mata MD, Ferraz-Santos B, Azevado ER, Giro EM, Zuanon AC. Addition of

chlorhexidine gluconate to a glass ionomer cement: a study on mechanical, physical and

antibacterial properties. Brazilian Dental Journal 2014; 25: 33-37.

11. Marti LM, Becci AC, Spolidorio DM, Brighenti FL, Giro EM, Zuanon AC. Incorporation

of chlorhexidine gluconate or diacetate into a glass-ionomer cement: porosity, surface

roughness, and anti-biofilm activity. American Journal of Dentistry 2014; 27: 318-22.

12. Takahashi Y, Imazato S, Kaneshiro AV, Ebisu S, Frencken JE, Tay FR. Antibacterial

effects and physical properties of glass-ionomer cements containing chlorhexidine for the ART

approach. Dental Materials 2006; 22: 647-52.

13. Cheng L, Weir MD, Xu HH, Kraigsley AM, Lin NJ, Lin-Gibson S, Zhou X. Antibacterial and

physical properties of calcium-phosphate and calcium-fluoride nanocomposites with

chlorhexidine. Dental Materials 2012; 28: 573-83.

14. Iz SG, Ertugrul F, Eden E, Gurhan SI. Biocompatibility of glass ionomer cements with

and without chlorhexidine. European Journal of Dentistry 2013; 7: S89-S93.

15. de Castilho AR, Duque C, Negrini Tde C, Sacono NT, de Paula AB, de Souza Costa CA,

Spolidorio DM, Puppin-Rontani RM. In vitro and in vivo investigation of the biological and

mechanical behaviour of resin-modified glass-ionomer cement containing chlorhexidine.

Journal of Dentistry 2013; 41: 155-63.

22

16. Barbour ME, Maddocks SE, Wood NJ, Collins AM. Synthesis, characterization, and

efficacy of antimicrobial chlorhexidine hexametaphosphate nanoparticles for applications in

biomedical materials and consumer products. International Journal of Nanomedicine 2013; 8:

3507-19.

17. Wood NJ, Jenkinson HF, Davis SA, Mann S, O'Sullivan DJ, Barbour ME. Chlorhexidine

hexametaphosphate nanoparticles as a novel antimicrobial coating for dental implants.

Journal of Materials Science: Materials in Medicine 2015; 26: 201.

18. Hook ER, Owen OJ, Bellis CA, Holder JA, O'Sullivan DJ, Barbour ME. Development of a

novel antimicrobial-releasing glass ionomer cement functionalized with chlorhexidine

hexametaphosphate nanoparticles. Journal of Nanobiotechnology 2014; 12: 3.

19. Barbour ME, O'Sullivan DJ, Jagger DC. Chlorhexidine adsorption to anatase and rutile

titanium dioxide. Colloids and Surfaces A 2007; 307: 116-20.

20. Bapna MS, Gadia CM, Drummond JL. Effects of aging and cyclic loading on the

mechanical properties of glass ionomer cements. European Journal of Oral Sciences 2002;

110: 330-34.

21. Mitra SB, Kedrowski BL. Long-term mechanical properties of glass ionomers. Dental

Materials 1994; 10: 78-82.

22. Baig MS, Dowling AH, Cao X, Fleming GJ. A discriminatory mechanical testing

performance indicator protocol for hand-mixed glass-ionomer restoratives. Dental Materials

2015; 31: 273-83.